I wanted a damaging spell for DnD 5e with a really high variance and decided to make my own. I think it’s reasonable, but I’d really love some peer review.

Mary’s Hail - 3rd level evocation

Casting Time: 1 action

Range: 60 ft

Components: V, S, M (a drop of holy water)

Duration: Instantaneous

Shards of holy ice hail down in a 5 ft diameter, 15 ft high cylinder. Each creature in the spell’s area must make a Dexterity saving throw or take 3d12 - 6 (minimum 0) radiant damage and 2d12 - 4 (minimum 0) cold damage and become coated in frost. You may use your bonus action to shatter the frost, dealing an additional 2d12 - 4 (minimum 0) cold damage to all affected creatures. After shattering, or at the end of your turn, the frost dissipates.

A few of my concerns:

  1. Is consistent access to bonus action usage too good?
  2. Will Sorcerer’s ability to re-roll dice break this spell?
  3. Is it too complicated?

The expected damage lines up with other damaging spells of 3rd level, balanced for:

  • average range, casting time, components

  • small aoe more for flavor than anything, but could be useful in some rare cases

  • no half damage on save

  • slightly reduced expected damage initially in exchange for the extra damage at the cost of a bonus action.

  • 3
    \$\begingroup\$ welcome! Just to make sure I'm reading correctly, the rolls each have damage subtracted on them? so, theoretically, the spell could deal zero damage? \$\endgroup\$
    – goodguy5
    Commented Aug 14, 2019 at 18:22
  • 5
    \$\begingroup\$ Negative damage is reduced to a minimum of 1 or 0? \$\endgroup\$
    – Blits
    Commented Aug 14, 2019 at 18:24
  • 1
    \$\begingroup\$ So, if I understand this spell, it can hit from 0 to ~27 targets (incorporating height and partial square coverage as hit-able), deals 0 - 30 radiant damage and 0 - 20 cold damage (for a total of 0-50 per creature, without bonus action), plus an additional 0-20 cold damage with the bonus action (for a total of 0-70 damage per target). Is that right? \$\endgroup\$
    – Upper_Case
    Commented Aug 14, 2019 at 18:32
  • 1
    \$\begingroup\$ Hi and welcome to RPG.SE! :) Take the tour if you haven't already. Could you please rephrase your question in such a way that it doesn't ask for general feedback? Instead of asking "what do you think of this?", try asking for a specific thing you are wondering about, e.g. how it compares to other 3rd-level spells regarding damage, complication, etc. This platform works best if you frame your question more clearly (the more specific, usually the better) while avoiding language that asks for general thoughts. Otherwise this Q is too much opinion-based. \$\endgroup\$
    – Vadruk
    Commented Aug 14, 2019 at 18:42
  • \$\begingroup\$ The difference between low damage and 0 damage is typically negligible, as players don't have access to HP knowledge to that accuracy. Why not go full in on the variance and use 1d20, 1d20, and 1d20, respectively, to eliminate the subtracted damage modifiers? Expected damage is 31.5 \$\endgroup\$
    – Blits
    Commented Aug 14, 2019 at 22:39

1 Answer 1


The spell is Mathematically Balanced; but Kludgy

Assuming the spellcaster makes consistent use of their Bonus Action to trigger the final frost damage, the spell will deal, on average (before Saving Throw mitigation) 31.564 damage. That's slightly more than a Fireball (28) but in a much smaller radius than a Fireball, so on that basis, I'm prepared to argue that this is appropriate for the level.

It should be acknowledged that Fireball is often considered "Overpowered" for its level, but we're not going to adjudicate that here.

I would argue, however, that the spell is awkward to use, which is probably a much bigger issue.

First off, the damage rolls of "3d12-6" and "2d12-4" are both awkward to work with because of the additional subtraction. It's not unreasonable, but it does make the spell stand out as an obvious departure from the common design principles of 5e. Adding or subtracting extra values doesn't really affect the variance of the damage that much (it'll actually reduce the variance a little because negative values will truncate to 0) but it makes the spell more awkward. So I would probably recommend just setting those values to 2d12 and 1d12 respectively.

Secondly, the way you've defined your area of effect is a bit unusual. There are very few spells (none I can specifically recall) that define a circular area by their diameter; instead, for circular areas, they're defined by their radius. With a 5 foot diameter, only a single medium-sized creature could be fit inside the area; or two creatures if they're both half-way inside the region. Most spells intended to fit this size of a region are usually just defined as a five foot cube, like with the Healing Spirit spell (Xanathar's Guide to Everything, pg. 157), or otherwise defined as a series of cubes of a certain size, like with the Fire Storm spell (Player's Handbook, pg. 242).

So my recommendation is to just allow the spell to target a cylinder with 5 foot radius, 15 feet tall. That increases the area of effect quite substantially, but that still makes it smaller than Fireball, and whether or not you decide to make the change to overall damage output, the spell should still feel appropriate for the level.

  • 1
    \$\begingroup\$ Thanks so much for the comprehensive feedback. I agree that the area is clunky, I’d probably change it to single-target. As for the dice, I’m aiming for inconsistency so dealing up to 70 dmg, with a non-negligible chance of dealing 0 damage fits better than 4 - 48 dmg. I agree that it makes the spell clunky and strange, but I think that’s what allows it to fill its niche. \$\endgroup\$ Commented Aug 14, 2019 at 20:24

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .